1 | // © 2016 and later: Unicode, Inc. and others. |
2 | // License & terms of use: http://www.unicode.org/copyright.html |
3 | /* |
4 | ********************************************************************** |
5 | * Copyright (C) 1999-2015, International Business Machines |
6 | * Corporation and others. All Rights Reserved. |
7 | ********************************************************************** |
8 | * file name: ustr_imp.h |
9 | * encoding: UTF-8 |
10 | * tab size: 8 (not used) |
11 | * indentation:4 |
12 | * |
13 | * created on: 2001jan30 |
14 | * created by: Markus W. Scherer |
15 | */ |
16 | |
17 | #ifndef __USTR_IMP_H__ |
18 | #define __USTR_IMP_H__ |
19 | |
20 | #include "unicode/utypes.h" |
21 | #include "unicode/utf8.h" |
22 | |
23 | /** |
24 | * Internal option for unorm_cmpEquivFold() for strncmp style. |
25 | * If set, checks for both string length and terminating NUL. |
26 | */ |
27 | #define _STRNCMP_STYLE 0x1000 |
28 | |
29 | /** |
30 | * Compare two strings in code point order or code unit order. |
31 | * Works in strcmp style (both lengths -1), |
32 | * strncmp style (lengths equal and >=0, flag true), |
33 | * and memcmp/UnicodeString style (at least one length >=0). |
34 | */ |
35 | U_CFUNC int32_t U_EXPORT2 |
36 | uprv_strCompare(const UChar *s1, int32_t length1, |
37 | const UChar *s2, int32_t length2, |
38 | UBool strncmpStyle, UBool codePointOrder); |
39 | |
40 | U_CAPI int32_t U_EXPORT2 |
41 | ustr_hashUCharsN(const UChar *str, int32_t length); |
42 | |
43 | U_CAPI int32_t U_EXPORT2 |
44 | ustr_hashCharsN(const char *str, int32_t length); |
45 | |
46 | U_CAPI int32_t U_EXPORT2 |
47 | ustr_hashICharsN(const char *str, int32_t length); |
48 | |
49 | /** |
50 | * Convert an ASCII-range lowercase character to uppercase. |
51 | * |
52 | * @param c A UChar. |
53 | * @return If UChar is a lowercase ASCII character, returns the uppercase version. |
54 | * Otherwise, returns the input character. |
55 | */ |
56 | U_CAPI UChar U_EXPORT2 |
57 | u_asciiToUpper(UChar c); |
58 | |
59 | // TODO: Add u_asciiToLower if/when there is a need for it. |
60 | |
61 | /** |
62 | * NUL-terminate a UChar * string if possible. |
63 | * If length < destCapacity then NUL-terminate. |
64 | * If length == destCapacity then do not terminate but set U_STRING_NOT_TERMINATED_WARNING. |
65 | * If length > destCapacity then do not terminate but set U_BUFFER_OVERFLOW_ERROR. |
66 | * |
67 | * @param dest Destination buffer, can be NULL if destCapacity==0. |
68 | * @param destCapacity Number of UChars available at dest. |
69 | * @param length Number of UChars that were (to be) written to dest. |
70 | * @param pErrorCode ICU error code. |
71 | * @return length |
72 | */ |
73 | U_CAPI int32_t U_EXPORT2 |
74 | u_terminateUChars(UChar *dest, int32_t destCapacity, int32_t length, UErrorCode *pErrorCode); |
75 | |
76 | /** |
77 | * NUL-terminate a char * string if possible. |
78 | * Same as u_terminateUChars() but for a different string type. |
79 | */ |
80 | U_CAPI int32_t U_EXPORT2 |
81 | u_terminateChars(char *dest, int32_t destCapacity, int32_t length, UErrorCode *pErrorCode); |
82 | |
83 | /** |
84 | * NUL-terminate a UChar32 * string if possible. |
85 | * Same as u_terminateUChars() but for a different string type. |
86 | */ |
87 | U_CAPI int32_t U_EXPORT2 |
88 | u_terminateUChar32s(UChar32 *dest, int32_t destCapacity, int32_t length, UErrorCode *pErrorCode); |
89 | |
90 | /** |
91 | * NUL-terminate a wchar_t * string if possible. |
92 | * Same as u_terminateUChars() but for a different string type. |
93 | */ |
94 | U_CAPI int32_t U_EXPORT2 |
95 | u_terminateWChars(wchar_t *dest, int32_t destCapacity, int32_t length, UErrorCode *pErrorCode); |
96 | |
97 | /** |
98 | * Counts the bytes of any whole valid sequence for a UTF-8 lead byte. |
99 | * Returns 1 for ASCII 0..0x7f. |
100 | * Returns 0 for 0x80..0xc1 as well as for 0xf5..0xff. |
101 | * leadByte might be evaluated multiple times. |
102 | * |
103 | * @param leadByte The first byte of a UTF-8 sequence. Must be 0..0xff. |
104 | * @return 0..4 |
105 | */ |
106 | #define U8_COUNT_BYTES(leadByte) \ |
107 | (U8_IS_SINGLE(leadByte) ? 1 : U8_COUNT_BYTES_NON_ASCII(leadByte)) |
108 | |
109 | /** |
110 | * Counts the bytes of any whole valid sequence for a UTF-8 lead byte. |
111 | * Returns 0 for 0x00..0xc1 as well as for 0xf5..0xff. |
112 | * leadByte might be evaluated multiple times. |
113 | * |
114 | * @param leadByte The first byte of a UTF-8 sequence. Must be 0..0xff. |
115 | * @return 0 or 2..4 |
116 | */ |
117 | #define U8_COUNT_BYTES_NON_ASCII(leadByte) \ |
118 | (U8_IS_LEAD(leadByte) ? ((uint8_t)(leadByte)>=0xe0)+((uint8_t)(leadByte)>=0xf0)+2 : 0) |
119 | |
120 | #ifdef __cplusplus |
121 | |
122 | U_NAMESPACE_BEGIN |
123 | |
124 | class UTF8 { |
125 | public: |
126 | UTF8() = delete; // all static |
127 | |
128 | /** |
129 | * Is t a valid UTF-8 trail byte? |
130 | * |
131 | * @param prev Must be the preceding lead byte if i==1 and length>=3; |
132 | * otherwise ignored. |
133 | * @param t The i-th byte following the lead byte. |
134 | * @param i The index (1..3) of byte t in the byte sequence. 0<i<length |
135 | * @param length The length (2..4) of the byte sequence according to the lead byte. |
136 | * @return true if t is a valid trail byte in this context. |
137 | */ |
138 | static inline UBool isValidTrail(int32_t prev, uint8_t t, int32_t i, int32_t length) { |
139 | // The first trail byte after a 3- or 4-byte lead byte |
140 | // needs to be validated together with its lead byte. |
141 | if (length <= 2 || i > 1) { |
142 | return U8_IS_TRAIL(t); |
143 | } else if (length == 3) { |
144 | return U8_IS_VALID_LEAD3_AND_T1(prev, t); |
145 | } else { // length == 4 |
146 | return U8_IS_VALID_LEAD4_AND_T1(prev, t); |
147 | } |
148 | } |
149 | }; |
150 | |
151 | U_NAMESPACE_END |
152 | |
153 | #endif // __cplusplus |
154 | |
155 | #endif |
156 | |